System-level performance study of interference alignment in cellular systems with base-station coordination
29 Nov 2012-pp 1155-1160
TL;DR: The results have shown that interference alignment is underperforming with respect to the baseline using base station coordination, when a realistic number of base stations is considered.
Abstract: The performance of any transmission scheme is coupled with the receive strategy. Herein the behavior of transmissions based on interference alignment scheme is investigated under different receive strategies. Moreover, interference alignment is compared with different state-of-art transmission schemes under the assumption of intrabase station and inter-base station coordination. The performance of the above-mentioned techniques is assessed with a fully 3GPP compliant downlink LTE simulator, which gives a very realistic picture of the behavior in real systems. The results have shown that interference alignment is underperforming with respect to the baseline using base station coordination, when a realistic number of base stations is considered. Moreover, the gains of interference alignment are also highly dependent on the receiver type, whereas the baselines show relatively low differences with different receivers.
Citations
More filters
TL;DR: In this article, it was shown that the high-power spectral efficiency is upper bounded by a quantity that does not depend on the transmit powers, and that cooperation is possible only within clusters of limited size, which are subject to out-of-cluster interference whose power scales with that of the incluster signals.
Abstract: Cooperation is viewed as a key ingredient for interference management in wireless networks. This paper shows that cooperation has fundamental limitations. First, it is established that in systems that rely on pilot-assisted channel estimation, the spectral efficiency is upper-bounded by a quantity that does not depend on the transmit powers; in this framework, cooperation is possible only within clusters of limited size, which are subject to out-of-cluster interference whose power scales with that of the in-cluster signals. Second, an upper bound is also shown to exist if the cooperation extends to an entire (large) system operating as a single cluster; here, pilot-assisted transmission is necessarily transcended. Altogether, it is concluded that cooperation cannot in general change an interference-limited network to a noise-limited one. Consequently, the existing literature that routinely assumes that the high-power spectral efficiency scales with the log-scale transmit power provides only a partial characterization. The complete characterization proposed in this paper subdivides the high-power regime into a degree-of-freedom regime, where the scaling with the log-scale transmit power holds approximately, and a saturation regime, where the spectral efficiency hits a ceiling that is independent of the power. Using a cellular system as an example, it is demonstrated that the spectral efficiency saturates at power levels of operational relevance.
363 citations
17 Oct 2014
TL;DR: Analytical expressions that enable quantifying the spectral efficiency of interference alignment (IA) in cellular networks without the need for simulation are presented and the benefits of IA are characterized.
Abstract: Capitalizing on the analytical potency of stochastic geometry and on some new ideas to model intercell interference, this paper presents analytical expressions that enable quantifying the spectral efficiency of interference alignment (IA) in cellular networks without the need for simulation. From these expressions, the benefits of IA are characterized. Even under favorable assumptions, IA is found to be beneficial only in very specific and relatively infrequent network situations, and a blanket utilization of IA is found to be altogether detrimental. Applied only in the appropriate situations, IA does bring about benefits that are significant for the users involved but relatively small in terms of average spectral efficiency for the entire system.
33 citations
Cites result from "System-level performance study of i..."
...In contrast with some prior works on the system-level performance of IA, which relied on simulations over grid networks [7]–[10], we set out to address the matter analytically in order to attain broader generality and more pronounced guidance in the conclusions....
[...]
TL;DR: The spectral efficiency achievable by interference alignment in a K-user multiple-input-multiple-output interference channel is studied in the face of time-selective continuous fading explicitly estimated through pilot-symbol observations and the robustness of IA in such operationally relevant conditions is assessed.
Abstract: The spectral efficiency achievable by IA (interfer- ence alignment) in a K-user MIMO (multiple-input multiple- output) interference channel is studied in the face of time- selective continuous fading explicitly estimated through pilot- symbol observations. The robustness of IA in such operationally relevant conditions is assessed through a joint optimization of the pilot overhead and the IA update interval, which are characterized—in high-power conditions—as solutions of a fixed- point equation. Variations of the formulation are given for both FDD (frequency-division duplexing) and TDD (time-division duplexing), the former requiring explicit feedback of the fading estimates and the latter relying on fading reciprocity. For the FDD variation, analog feedback is considered. In addition to arbitrary numbers of users and antennas, and arbitrary tem- poral fading correlation functions, the derivations accommodate forward and reverse links with asymmetric power levels.
28 citations
Cites background from "System-level performance study of i..."
...Tempered by the considerations made in [33]–[35], chiefly that the insights obtained from the K-user interference channel apply to large wireless networks only within a certain SNR range, IA is seen to retain the potential to play some role in the management of interference in pedestrian-oriented TDD wireless systems; that role, however, might be largely circumscribed to cell-edge users for which the interference channel is a reasonable model....
[...]
TL;DR: A generalized iterative algorithm which directly maximizes the sum-rate without assuming the signal-to-noise ratio to be infinite is proposed, and a new design approach based on weighted-sum-rate maximization assuming a virtual equalizer type at the transmitter to limit the optimization process to the transmitter side.
Abstract: In this paper, we study the problem of per-stream maximum sum-rate joint precoder and minimum mean-squared error equalizer design for the multi-input multi-output interference channel. We consider the general case of more than three users with more than one stream per user. We propose a generalized iterative algorithm which directly maximizes the sum-rate without assuming the signal-to-noise ratio to be infinite. To reduce complexity, which can become prohibitive for large network size, we examine the performance-complexity tradeoffs involved in a sparse equalizer design. Joint precoder and equalizer optimization requires alternation between the forward and reverse links and assumes perfect synchronization between the transmitters and receivers at each network node, resulting in extensive overhead and spectral efficiency loss. To overcome this serious drawback, we propose a new design approach based on weighted-sum-rate maximization assuming a virtual equalizer type at the transmitter to limit the optimization process to the transmitter side. In addition, we quantify the sum-rate loss due to mismatched equalizer types and demonstrate the robustness of our proposed sum-rate weighting strategy to such mismatches with perfect or imperfect channel knowledge. Finally, we derive asymptotic performance expressions and verify their accuracy numerically even for a moderate number of users.
23 citations
01 Nov 2012
TL;DR: This paper studies the fundamental performance of IA, in the context of a large cellular network, and contrasts it with that of non-cooperative MIMO.
Abstract: Distributed cooperation schemes such as Interference Alignment hold the promise of an increased number of spatial degrees of freedom and, with that, of substantially higher spectral efficiencies. Most results available to date, however, have been obtained in simplified settings featuring a small number of transmitters and receivers in isolation. While such controlled settings are excellent platforms to develop ideas and build intuition, they also conceal important aspects that are inherent to actual wireless systems. Chief among these is the fact that any small set of cooperating transmitters and receivers is bound to be embedded within a large system featuring many other transmitters and receivers. This paper studies the fundamental performance of IA, in the context of a large cellular network, and contrasts it with that of non-cooperative MIMO.
16 citations
References
More filters
TL;DR: For the fully connected K user wireless interference channel where the channel coefficients are time-varying and are drawn from a continuous distribution, the sum capacity is characterized as C(SNR)=K/2log (SNR)+o(log( SNR), which almost surely has K/2 degrees of freedom.
Abstract: For the fully connected K user wireless interference channel where the channel coefficients are time-varying and are drawn from a continuous distribution, the sum capacity is characterized as C(SNR)=K/2log(SNR)+o(log(SNR)) . Thus, the K user time-varying interference channel almost surely has K/2 degrees of freedom. Achievability is based on the idea of interference alignment. Examples are also provided of fully connected K user interference channels with constant (not time-varying) coefficients where the capacity is exactly achieved by interference alignment at all SNR values.
3,385 citations
TL;DR: This paper proposes designing precoders by maximizing the so-called signal-to-leakage-and-noise ratio (SLNR) for all users simultaneously, and it also avoids noise enhancement.
Abstract: In multiuser MIMO downlink communications, it is necessary to design precoding schemes that are able to suppress co-channel interference. This paper proposes designing precoders by maximizing the so-called signal-to-leakage-and-noise ratio (SLNR) for all users simultaneously. The presentation considers communications with both single- and multi-stream cases, as well as MIMO systems that employ Alamouti coding. The effect of channel estimation errors on system performance is also studied. Compared with zero-forcing solutions, the proposed method does not impose a condition on the relation between the number of transmit and receive antennas, and it also avoids noise enhancement. Simulations illustrate the performance of the scheme
871 citations
"System-level performance study of i..." refers methods in this paper
...First, the performance of IA is compared with the state-of-art baseline [7] using a coordinated beamforming scheme based on signal to leakage and noise ratio (SLNR) optimization....
[...]
...2) Coordinated Beamforing via (SLNR) We consider a state-of-art coordinated beamforming (CB) scheme based on a SLNR criterion [7]....
[...]
08 Dec 2008
TL;DR: Examples of iterative algorithms that utilize the reciprocity of wireless networks to achieve interference alignment with only local channel knowledge at each node are provided, providing numerical insights into the feasibility of interference alignment that are not yet available in theory.
Abstract: Recent results establish the optimality of interference alignment to approach the Shannon capacity of interference networks at high SNR. However, the extent to which interference can be aligned over a finite number of signalling dimensions remains unknown. Another important concern for interference alignment schemes is the requirement of global channel knowledge. In this work we provide examples of iterative algorithms that utilize the reciprocity of wireless networks to achieve interference alignment with only local channel knowledge at each node. These algorithms also provide numerical insights into the feasibility of interference alignment that are not yet available in theory.
859 citations
"System-level performance study of i..." refers background or methods in this paper
...1) Interference Alignment (IA) The IA based transmit precoding scheme [1],[2] adjusts the interference space received at the ith UE, such that UE ith’s receiver has enough degrees-of-freedom to cancel this interference....
[...]
...We refer to [2] for a detailed description of the IA concept....
[...]
...Initial works (see for example [1], [2] and references wherein) studied the behavior of IA under ideal assumptions, like Gaussian i....
[...]
...The receiver weights in [2] are calculated in order to satisfy the following conditions...
[...]
TL;DR: In this paper, the authors used analytical and computer simulation techniques to determine the performance of optimum combining when the received desired and interfering signals are subject to Rayleigh fading, and they showed that optimum combining is significantly better than maximal ratio combining even when the number of interferers is greater than number of antennas.
Abstract: This paper studies optimum signal combining for space diversity reception in cellular mobile radio systems. With optimum combining, the signals received by the antennas are weighted and combined to maximize the output signal-to-interference-plus-noise ratio. Thus, with cochannel interference, space diversity is used not only to combat Rayleigh fading of the desired signal (as with maximal ratio combining) but also to reduce the power of interfering signals at the receiver. We use analytical and computer simulation techniques to determine the performance of optimum combining when the received desired and interfering signals are subject to Rayleigh fading. Results show that optimum combining is significantly better than maximal ratio combining even when the number of interferers is greater than the number of antennas. Results for typical cellular mobile radio systems show that optimum combining increases the output signal-to-interference ratio at the receiver by several decibels. Thus, systems can require fewer base station antennas and/or achieve increased channel capacity through greater frequency reuse. We also describe techniques for implementing optimum combining with least mean square (LMS) adaptive arrays.
621 citations
"System-level performance study of i..." refers background in this paper
...3) Type 3: MMSE with instantaneous interfeence covariance matrix estimation (MMSE-I) The well known Minimum Mean Square Error (MMSE) receiver based on an instantaneous estimation of the interference covariance matrix can be written as [8]...
[...]
19 May 2008
TL;DR: It is shown that the sum capacity of the K user frequency selective (or time-varying) interference channel is C(SNR) = (K/2) log( SNR) +o(log(SN R)) meaning that the channel has a total of K/2 degrees of freedom per orthogonal time and frequency dimension.
Abstract: We show that the sum capacity of the K user frequency selective (or time-varying) interference channel is C(SNR) = (K/2) log(SNR) +o(log(SNR)) meaning that the channel has a total of K/2 degrees of freedom per orthogonal time and frequency dimension. Linear schemes of interference alignment and zero forcing suffice to achieve all the degrees of freedom and multi-user detection is not required.
592 citations